Recording Bangs - Hardware

[dbsweng]

Hello!

Sorry, this might be a bit off topic but I don’t even know what to search for.

I have a microphone recording 24/7 and I’m trying to detect bangs in the resulting files (one hour slices). When I do find the bangs, they are poorly recorded. I am told that this is because of Automatic Gain Control not being able to adjust in time.

Does anyone know much about this topic? or can anyone direct me to a forum for people who would have an interest in that?

[Trebor]

@dbsweng

Automatic Gain Control …

I think you mean sound activated recording,
if you’re recording using Audacity that can be switched off …
https://manual.audacityteam.org/man/recording_preferences.html

@dbsweng

I’m trying to detect bangs in the resulting files

Here’s a similar question …

Trying To Record Bass Sounds, Thumps, Thuds.

(rhythmic) bassy thumps are easier to see on a spectrogram than to reproduce … [bassy thumps can be seen on the spectrogram (20-200Hz).png] The spectrogram could be evidence that the thumping occurred at particular time, but it’s not evidence of how loud the thumping was: playback could be louder than the original sound.

[dbsweng]

Hey Trevor, thanks for the reply

I don’t think that’s what I want (but I might be wrong, my knowledge of audio recording is almost zilch).

Imagine you have a microphone in an empty room. You record for one hour, and you’re trying to determine if a door banged one or more times. There’s no other noise being generated inside the room on purpose (i.e. there’s clicks from hard drives and so on, but no conversations or moving objects).

In the recording there is a bang, but the magnitude of the bang is far below what was audible to the human ear. Another person has told me that the AGC of the microphone (not audacity) may have caused the bang to be recorded at a lower volume.

I need to know how to fix that. That is to say, I want to record the bang at the correct volume.

[Trebor]

@dbsweng

I don’t think that’s what I want

Sorry I thought something was preventing you from establishing the time that the recorded banging noise occurred.

@dbsweng

… banged one or more times. There’s no other noise being generated inside the room on purpose …

Before you call an exorcist,
you, (or your neighbors), could have water-hammer.

Freesound - Water_Hammer_01.wav by dheming

[dbsweng]

Ahhh water hammer, a distant relative of mc hammer.

Sorry Trebor I’ve got more beginner questions.

1 - what hardware did you record that on? it seems to have recorded the bang quite well.

2 - Was the bang also emanating from the bathroom sink or was the bathroom sink merely further away from the source?

I’m actually not looking for a poltergeist (yet - that may change if this project is successful). I actually can’t remember why I got so interested in this, I think there was just a noteworthy bang that was only slightly recorded by the mic and I was curious about that.

[Trebor]

@dbsweng

what hardware did you record that on?

I didn’t record that example of water hammer, (which is a mild example: it can sound like someone banging on a door, and be caused by neighbor’s turning off water).

If you cannot switch the AGC off on your recorder,
you could keep its gain constant by putting it on something which makes a constant sound, like a ticking (mechanical) alarm clock.

[dbsweng]

Ahhh!

Oh that’s a good idea! Thank you!

[dbsweng]

For any lost soul who wanders upon this place…

The ticking didn’t work how I wanted it to. That registered nicely on the audio but the sudden impulse bangs just sounded like mild pops still.

I ended up using a condenser microphone: Behringer BIGFOOT All-In-One USB Studio Condenser Microphone

This isn’t an Audacity specific thing, so thread continues here: Detecting Bangs - Raspberry Pi Forums